I have some new boards, but the parts supply has disappeared before I could stock up. My apologies - I am preparing a revised circuit, returning to all-discrete, for ease of procurement!
I think it is worth building it as a real circuit, and measuring it up.
One deception that SPICE always plays on us is to give results that you would get if all the triodes (& transistors..) were exact matches. This is an important consideration for your circuit, as the two halves of the differential pair will degrade the overall performance, with real-world mismatch.
But you can reduce the errors by adding small-value resistors between the CCS tail, and each cathode of U4a and U4b. Try the sims again with say, 68Ω and 100Ω. higher values give better match, but lower gain; so as always you must find the right compromise.
Last edited:
That is good advice for sure. Thanks.
Before getting into real-world prototyping, since this is the first time I'm not going with a fully pre-made design, I wanted to get a few eyes on the starting point (just being the driver stage in this case) to check whether there obvious problems with what I've cooked up so far...
Also tube type recommendation would be interesting.
Has anyone used the 6C6 as a driver for a 300B? I have a bunch of them as well as two pair of EML 300bs.
I've been trying out 6C4C (Russian 6B4G) as a preamp tube and like it a lot. So my very left field suggestion would be a 3 way design using 6C4C into whatever, even a plain old 6SN7 driver. I want to build this myself - very curious. My usual circuits are 2 stage.